186 Helene C. Muller-Landau
likely to win the site. Geritzet al. (1999) consider
the evolutionary as well as ecological dynamics
of this model, and show that as the degree of
competitive asymmetry increases, the number
of types (species) that evolve and stably coexist
increases. Adler and Mosquera (2000) analyt-
ically derive the conditions under which one,
two, and infinite numbers of species can coexist
via the competition–colonization trade-off, specif-
ically showing that infinite coexistence is possible
only under perfect asymmetry. Kisdi and Geritz
(2003a) demonstrate similar effects of varying
asymmetry in models of perennial plants.The role
of asymmetry in these models is consistent with
the results of Tilman (1994) and Kisdi and Geritz
(2003b), who find coexistence of infinite numbers
of species in models with perfectly asymmetric
competition, and of Levine and Rees (2002) who
find limited coexistence under low asymmetry.
Thus, the classical competition–colonization
trade-off can be a strong stabilizing force for
diversity maintenance, but only if there is suffi-
cient competitive asymmetry. Perfect asymmetry,
which is unrealistic for real communities (Adler
and Mosquera 2000), is required for the effec-
tively infinite coexistence attained in the original
theoretical models (Tilman 1994). In contrast,
if competition is perfectly symmetric, then the
contribution of this trade-off alone to diversity
maintenance can be only equalizing.The perfectly
equalizing case is essentially infinitely unlikely to
occur; even small deviations that make the trade-
off partially rather than perfectly equalizing are
sufficient to make some species superior com-
petitors and shorten coexistence (Zhang and Lin
1997, Yuet al. 1998). Finally, if competition is
partially asymmetric, the most likely case in real
communities, then the trade-off may be able to
contribute to stable coexistence of a few species,
or it may be merely a partially equalizing force.
COLONIZATION-RELATED
TRADE-OFFS AND HABITAT
PARTITIONING
While only certain competition–colonization
trade-offscanbeastabilizinginfluenceondiversity
maintenance in homogeneous environments, a
wider range of colonization-related trade-offs can
have stabilizing influences given appropriate spa-
tial or temporal environmental heterogeneity.
Specifically, these trade-offs can contribute to
diversity maintenance if the combination of
each species’ colonization and competitive abil-
ities on the different habitats is such that each
species has the highest population growth rate in
some time or place (Chesson and Warner 1981,
Comins and Noble 1985, Yu and Wilson 2001).
Because both habitat heterogeneity and varia-
tion in species performance on different habitats
are ubiquitous in real ecosystems, these trade-
offs have the potential to play important roles
in diversity maintenance. Here I consider two
specific examples – tolerance–fecundity trade-offs
and dispersal–fecundity trade-offs.
A tolerance–fecundity trade-off can mediate
coexistence when there is spatial variation in
resourceavailabilityandthusinthelevelof recruit
provisioning needed to tolerate local conditions
and have a chance at winning the regeneration
site. In this case, a trade-off between recruit provi-
sioning (e.g., seed mass) and fecundity (e.g., seed
production) can mediate coexistence by allowing
the more fecund species to succeed dispropor-
tionately often in sites where little provisioning
is needed, and thus make u pfor the consistent
success of the better-provisioned species on sites
where resource availability is low (Levine and
Rees 2002). In principle, many species can coexist
given sufficient variation in habitat quality among
sites, and appropriate consistency in the trade-off
between habitat tolerance and fecundity. Specif-
ically, such a trade-off will be stabilizing if the
fecundity of each less tolerant species exceeds that
of the next more tolerant species by a particular
multiple, with that multiple depending on their
relative habitat tolerances, and seed survival, if
relevant. If the fecundity of the less tolerant
species is less than (or equal to) this multiple
of the fecundity of the more tolerant species,
then the trade-off will be partially (or perfectly)
equalizing.
A dispersal–fecundity trade-off can allow two
competitors to coexist given spatial variation in
the density of potential regeneration sites (Yu and
Wilson 2001). The more fecund species is more
successful in areas of high site density and the